Journal of Neurochemistry, 2003, 87, 437–447

doi:10.1046/j.1471-4159.2003.02012.x

Induction of an interferon-c Stat3 response in nerve cells by pre-treatment with gp130 cytokines Navjot Kaur, In-Jung Kim, Dennis Higgins and Stanley W. Halvorsen Department of Pharmacology and Toxicology, School of Medicine and Biomedical Sciences, The State University of New York at Buffalo, Buffalo, New York, USA

Abstract Many cytokines mediate their effects through Jak/STAT signaling pathways providing many opportunities for cross-talk between different cytokines. We examined the interaction between two cytokine families, gp130-related cytokines and interferon-c (IFN-c), which are coexpressed in the nervous system during acute trauma and pathological conditions. Typical nerve cells show an IFN-c response that is restricted to activating STAT1, with minor activation of STAT3. IFN-c elicited a pronounced STAT3 response in cells pre-treated for 5–7 h with ciliary neurotrophic factor (CNTF), leukemia inhibitory factor or interleukin-6. CNTF or interleukin-6 induced an IFN-c STAT3 response in a variety of cells including SH-SY5Y human neuroblastoma, HMN-1 murine motor neuron hybrid cells, rat sympathetic neurons and human hepatoma HepG2 cells. The enhancement was measured as

an increase in tyrosine phosphorylated STAT3, in STAT3DNA binding and in STAT-luciferase reporter gene activity. The enhanced STAT3 response was not due to an increase in overall STAT3 levels but was dependent upon ongoing protein synthesis. The induction by CNTF was inhibited by the protein kinase C inhibitor, BIM, and the MAPK-kinase inhibitor, U0126. Further, H-35 hepatoma cells expressing gp130 receptor chimeras lacking either the SHP-2 docking site or the Box 3 STAT binding sites failed to enhance the IFN-c STAT3 response. These results provide evidence for an interaction between gp130 and IFN-c cytokines that can significantly alter the final cellular response to IFN-c. Keywords: ciliary neurotrophic factor, MAPK kinase, neuroblastoma cells, neurotrophic factor, SHP-2, signal transduction. J. Neurochem. (2003) 87, 437–447.

Cytokines regulate critical cellular processes including proliferation, development and survival (Kotzbauer et al. 1994) as well as playing roles in host defense and immunopathological processes (Leonard and Oshea 1998). Many cytokines mediate these important functions through a Jak/ STAT signaling pathway (Darnell et al. 1994). They do so by activating selective members of the Jak kinase and STAT transcription factor families. However, cytokines, at times, also have the capacity to activate alternative or even multiple signaling pathways and these pathways are shared among different cytokines and growth factors (Stahl et al. 1995; Taga 1996). The various mechanisms by which cells regulate the relative activity through different pathways are still not well understood but are likely to result from the different environments and histories of cells. The cytokine, interferon-c (IFN-c), activates a Jak/STAT pathway through its receptor a- and b-subunits (Stark et al. 1998). In almost all cells IFN-c binds to its a- and b-subunits leading to activation of Jak1 and Jak2 kinases followed by tyrosine phosphorylation of STAT1 (Shuai et al. 1993;

Boehm et al. 1997; Stark et al. 1998). In rare instances IFN-c also activates STAT3, but this response is restricted to certain cell types (Stephens et al. 1998; Caldenhoven et al. 1999; Hu et al. 2002; Orlovsky et al. 2002). Cytokines such

Received April 30, 2003; revised manuscript received June 24, 2003; accepted July 5, 2003. Address correspondence and reprint requests to Stanley W. Halvorsen, Department of Pharmacology and Toxicology, 102 Farber Hall, University at Buffalo, Buffalo, NY 14214–3000, USA. E-mail: [email protected] Abbreviations used: BIM, bisindolylmaleimide-I; BMP, bone morphogenetic protein; CNTF, ciliary neurotrophic factor; ECL, enhanced chemiluminescence assay; EMSA, electrophoretic mobility shift assay; G-CSF, granulocyte colony stimulating factor; gp130, glycoprotein 130 subunit of the CNTF receptor; IFN-c, interferon-c; IgG-HRP, horse radish peroxidase linked immunoglobulin G; IL-6, interleukin-6; Jak, Janus kinase; LIF, leukemia inhibitory factor; MAPK, mitogen-activated protein kinase; PI-3 kinase, phosphatidyl inositol-3-kinase; (PY)-STAT, (tyrosine phosphorylated)-signal transducer and activator of transcription; SH2, src homology domain 2; SHP-2, SH2-containing protein tyrosine phosphatase-2.


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

437

438 N. Kaur et al.

as ciliary neurotrophic factor (CNTF), leukemia inhibitory factor (LIF), oncostatin M, cardiotrophin-1, cardiotrophinlike cytokine, interleukin-6 (IL-6) and interleukin-11 act through a common signal-transducing receptor subunit, glycoprotein 130 (gp130), in addition to a variable b subunit (Stahl et al. 1994; Taga 1996). These CNTF-like cytokines bind to their receptors to activate Jak1 and Jak2 resulting in the tyrosine phosphorylation and activation preferentially of STAT3, and to a lesser extent, STAT1 (Stahl et al. 1995; Taga 1996; Wishingrad et al. 1997). Other signal transduction components can also be activated by CNTF-like cytokines in different cells and under different conditions including phosphatidyl inositol-3-kinase (PI-3 kinase), mitogen-activated protein kinase (MAPK) and protein tyrosine phosphatase (Stahl et al. 1994; Taga 1996). Phosphorylated STAT1 and/or STAT3 homo- or heterodimerize and translocate to the nucleus to interact with specific DNA response elements and regulate the expression of the targeted genes (Shuai et al. 1993; Stahl et al. 1994; Ihle 1996; Wells and deVos 1996). Although CNTF and IFN-c share elements of the same Jak/STAT signaling pathway, their physiological functions are different and when examined independently they induce different sets of genes (Yuan et al. 1994). IFN-c regulates the immune system and the response to infectious agents (Boehm et al. 1997; Leonard and Oshea 1998). The physiological functions of CNTF are still under study, but it (and related factors) has profound effects on neuronal phenotype and development of the nervous system (Ernsberger et al. 1989; Halvorsen and Berg 1989; Symes et al. 1994; Finn et al. 1998) and is involved in responses to traumatic nerve damage (Sendtner et al. 1990). Indeed, CNTF and LIF are both released in significant amounts at sites of nerve damage (Winter et al. 1995; Kurek et al. 1996; Sun et al. 1996). Other members of the LIF/CNTF family are involved in hematopoiesis (Metcalf 2003), cardiac hypertrophy (Hirota et al. 1995; Pennica et al. 1995), the acute phase response and proinflammatory responses (Levison et al. 1996). Therefore it seems inevitable that cells will be exposed to combinations of these cytokines at various stages of development or during pathological conditions. The outcomes, nature and mechanisms of these interactions are a growing area of interest. Activated STAT1 and STAT3 have been shown to bind the common regulatory elements of several different genes, but they also can selectively interact with distinct and specific elements in separate genes (Kordula et al. 1995; Lamb et al. 1995; Mahboubi and Pober 2002). The response of a cell will be determined by the complement of STATs active at any time. Thus, a tight balance between multiple signaling pathways must be maintained to carry out specific functions in the cells. We are interested in understanding the role of these signaling pathways in specific cellular functions and how these pathways are

regulated under developmental and environmental conditions. In earlier studies we found that continuous exposure of cells to CNTF-like cytokines causes desensitization of all members of the gp130 cytokine family (Kaur et al. 2002). In this study, we show that continuous exposure of cells to CNTF and related gp130 cytokines greatly enhances the STAT3 response of cells to IFN-c while leaving the STAT1 response unaffected. Using a combination of molecular studies with mutated chimeric gp130 receptors and pharmacologic inhibitors, we find that this cross-talk requires CNTF activation of both STAT1 and MAPK pathways.

Materials and methods Materials Human recombinant CNTF was provided by Regeneron Pharmaceuticals (Tarrytown, NY, USA), LIF and IL-6 were from R & D Systems (St. Paul, MN, USA) and human IFNc was purchased from Peprotech International (Rocky Hill, NJ, USA). Bisindolymaleimide I (BIM), LY294002, U1026 and H7 were from Calbiochem (La Jolla, CA, USA). The monoclonal anti-STAT antibodies were from Transduction Laboratories (Lexington, KY, USA) and antiphosphotyrosine STAT (PY-STAT) antibodies were from New England Biolabs (Beverly, MA). Anti-phosphotyrosineJak (PY-Jak) antibodies were from Biosource International (Camarillo, CA, USA) and antibodies to Jak were from Santa Cruz Biotechnologies (Santa Cruz, CA, USA). Cell culture reagents, goat anti-mouse IgG-HRP antibodies and Lipofectamine 2000 were purchased from Invitrogen (Carlsbad, CA, USA) and goat anti-mouse IgG-HRP was from Cappel Laboratories (Durham, NC, USA). Cell culture and drug treatment The human neuroblastoma cell line, SH-SY5Y, was obtained from Dr June Biedler of the Sloan Kettering Institute for Cancer Research. SH-SY5Y cells were grown in a 1 : 1 mixture of Ham’s F12 and Eagle’s minimal essential medium supplemented with 10% (v/v) fetal bovine serum, 50 U/mL penicillin and 50 lg/mL streptomycin as previously described (Malek and Halvorsen 1997). Hybrid, HMN-1 cells (Salazar-Grueso et al. 1991; Zhou et al. 1998), derived from embryonic mouse spinal cord motor neurons and neuroblastoma cells, were obtained from R. Pittman (University of Pennsylvania, Philadelphia, PA, USA) and were grown in the same conditions as SH-SY5Y cells but supplemented with 5% (v/v) fetal bovine serum. Cells were placed in serum-free medium for 1–5 h before initiating drug treatments. Pre-treatment of cells were performed with 1 nM CNTF for 5 h and acute challenges with cytokines (100–200 pM) were given for 30–45 min unless otherwise stated. The human hepatoma cell line, HepG2, and the rat H-35 cell lines stably expressing chimeric G-CSF/gp130 receptors with mutations in their cytoplasmic domains were generously provided by H. Baumann (Roswell Park Cancer Institute, Buffalo, NY, USA). These chimeric receptors have the extracellular domain of the granulocyte-colony stimulating factor (G-CSF) receptor and the membrane proximal 133 amino acid residues of the intracellular


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

CNTF induction of interferon-c STAT3 signaling 439

domain of gp130. The truncated receptor, G-CSF/gp130(133)WT, retains a single Box3 STAT binding site, Jak binding and the SHP-2 recruitment site and is capable of activating Jak, STAT, SHP-2 and Erk1/2. The second cell line, G-CSF/gp130(133)Y2F, expresses a receptor deficient in SHP-2 and Erk1/2 activation due to a Tyr to Phe mutation at position 117 but retains Jak and STAT activation. The third cell line, G-CSF/gp130(133)Y3F, expresses a receptor lacking all STAT binding sites but retains the SHP-2 site (Blanchard et al. 2001). Superior cervical ganglia were dissected from perinatal (embryonic day 21) Holtzman rats (Harlan–Sprague–Dawley, Rockford, IL, USA) as previously described (Higgins et al. 1991). After dissociation with trypsin (2.5 mg/mL) and collagenase (1 mg/mL) for 40 min, cells were plated onto poly D-lysine coated (100 lg/mL) 35-mm dishes. Cells were maintained in serum-free medium containing b-nerve growth factor (100 ng/mL). Then 1–2 days later, cytosine-b-D-arabinofuranoside (1 lm) was added to the medium for 2 days to kill non-neuronal cells. The cultures were allowed to recover for 1 day and experimental treatments were begun on the sixth or seventh day. Immunoblot analysis Cell monolayers were harvested in SDS–polyacrylamide gel sample buffer and proteins were separated on 7.5% gel before transfer onto polyvinylidene difluoride membranes. Immunodetection of proteins on immunoblots was performed as described previously using Amersham ECL-Plus (Kaur et al. 2002). Exposed films were scanned with an Epson 636 Professional Series scanner and analyzed using the public domain program NIH image. Data are presented as the mean ± SEM. CNTF induction of the PY-STAT3 response to IFN-c was quantified by calculating the ratio of the IFN-c response after pre-treatment to the IFN-c response without CNTF pre-treatment less the CNTF pre-treatment value. In data sets where there were experiments with no PY-STAT3 IFN-c response without pre-treatment, the induction was calculated by the ratio of the IFN-c response after CNTF pre-treatment alone plus the response to IFN-c 30 min alone (e.g. Figs 2 and 5). PY-STAT levels were normalized to levels of either STAT3 or STAT6 (as indicated) determined by stripping and reprobing each blot. Electrophoretic mobility shift assay Preparation of nuclear extracts and DNA binding assays were performed as previously described (Kaur et al. 2002) with minor changes. Briefly, nuclear extracts were incubated with 1 lg of poly(dI-dC) in a binding buffer containing 20 mM Hepes pH 7.9, 20 mM KCl, 1 mM MgCl2, 1 mM dithiothreitol, 0.5 mM EDTA and 10% glycerol for 2 h at 4C. Double-stranded high-affinity SIEoligonucleotide (m67; Geneka, Montreal, Quebec) from the c-fos promoter (5¢-GTC GAC ATT TCC CGT AAA TCG TCG A-3¢) labeled with c-32P-dATP was added and incubated for additional 20 min on ice. For DNA supershift assay, antibodies against STAT1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or STAT3 (Zymed, South San Francisco CA, USA) were included in the reaction 2 h prior to the addition of labeled probe. The complexes were resolved on a 4.5% non-denaturing polyacrylamide gel in 20 mM Tris/HCl (pH 8.0), 20 mM boric acid and 0.5 mM EDTA. Gels were fixed, dried and subjected to autoradiography.

Transient transfection and luciferase reporter expression assay HMN-1 cells were transfected using lipofectamine 2000 according to the manufacturer’s instructions. Briefly, cells were plated in 24well plates and 24 h later growth medium was replaced with 0.5 mL of serum-free media. To each well, 0.1 mL of a mixture containing 2 lL of lipofectamine and 1 lg 3(G)3 luciferase reporter plasmid or 0.5 lg of the cotransfection control, RSV gal were added. The reporter plasmid (provided by A. Symes, USUHS, Bethesda, MD, USA) contained three 20 bp copies of the VIP CyRE STAT3 site upstream from the RSV promoter driving luciferase (Symes et al. 1995). Cells were subsequently treated with cytokines beginning 24 h after transfection. After 36 h, cells were harvested and assayed for luciferase activity (Promega, Madison, WI, USA) and b-galactosidase activity as previously described (Malek and Halvorsen 1997). Luciferase activity was normalized to b-galactosidase activity as a control for transfection efficiency. HepG2 cells were transfected with 1 lg of a plasmid containing hemagglutinin tagged STAT wild type genes, HA-STAT1WT, HA-STAT3WT, or the dominant-negative Phe substituted Tyr701 or Tyr705 constructs, HA-STAT1F or HA-STAT3F, respectively, cloned into the pCAGGS expression vector (Nakajima et al. 1996). STAT1- and STAT3-containing plasmids were provided by Dr T. Hirano (Osaka University Medical School, Japan). To determine the transfection efficiency, 0.25 lg of a pEGFP-N2 plasmid was used (Clontech, Palo Alto, CA, USA). After 48 h the medium was changed and cells were allowed to recover for 2 h before treating with cytokines.

Results

CNTF pre-treatment increases IFN-c coupling to Jak/STAT3 Treatment of SH-SY5Y neuroblastoma cells with IFN-c for 30–45 min produced a maximal PY-STAT1 response at concentrations as low as 100 pM (Fig. 1). IFN-c concentrations greater than 100 pM were required to consistently elicit significant PY-STAT3 responses (Fig. 1b). Prolonged pretreatment of SH-SY5Y cells first with CNTF resulted in an induction of the IFN-c-mediated PY-STAT3 response with no change in the PY-STAT1 response (Fig. 1). Analysis of dose–response effects of IFN-c with and without CNTF pretreatment revealed a decrease in the EC50 value of cells for IFN-c by about 10–20-fold after CNTF pre-treatment while the sensitivity of cells to stimulation of PY-STAT1 by IFN-c was unchanged by CNTF pre-treatment (data not shown). The induction of the IFN-c PY-STAT3 response by CNTF pre-treatment was maximal after 30 min of IFN-c challenge, but could be observed at all times measured between 10 min and 2 h (Fig. 1c,d). Therefore, CNTF pre-treatment had a major effect on inducing or increasing the IFN-c-mediated PY-STAT3 response, without affecting the PY-STAT1 response. At the same time, the PY-STAT3 response to other gp130-related cytokines was completely prevented by this pre-treatment paradigm with CNTF (Kaur et al. 2002).


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

440 N. Kaur et al.

Fig. 1 CNTF induction of an IFN-c STAT3 response in SH-SY5Y cells. (a) Cultures of SH-SY5Y cells were treated with CNTF (1 nM) for 5 h or IFN-c (100 pM) for 30 min or CNTF (1 nM) for 5 h followed by IFN-c (100 pM) for 30 min. Cell lysates were collected and processed for immunoblot analysis by ECL. Blots were probed sequentially with anti-PY-STAT3 antibodies, anti-PY-STAT1 antibodies and anti-STAT3 antibodies. (b) SH-SY5Y cells were not pre-treated or pre-treated for 5 h with CNTF (1 nM) before challenge with 0, 0.1, 0.5, 1.0 and 3.0 nM IFN-c as indicated for 30 min and then processed for immunoblot analysis as in (a) using anti-PY-STAT1 and PY-STAT3 antibodies. (c,d) SH-SY5Y cells were not pre-treated (open symbols) or pretreated (closed symbols) with CNTF (1 nM) for 5 h and then challenged for the indicated times with IFN-c (100 pM). Following immunoblot analysis, PY-STAT3 (c) and PY-STAT1 (d) levels were normalized to total STAT3 and plotted as a percent of the maximum PY-STAT response for the non-CNTF pre-treatment group (n ¼ 3–4 determinations).

Concentration- and time-dependence of CNTF pre-treatment on induction of IFN-c STAT3 responses The extent of the induction of the IFN-c PY-STAT3 response was dependent on the concentration of CNTF in the pretreatment medium. Induction was increased at concentrations from 0.1 to 100 pM (Fig. 2a). The duration of CNTF pretreatment was also critical to induction of an IFN-c PY-STAT3 response. Pre-treatment with CNTF for 30–60 min was inadequate to reveal the induction response while durations of 3–5 h reliably induced the IFN-c-mediated PY-STAT3 response (Fig. 2b). SH-SY5Y cells pre-treated with another gp130 cytokine, LIF, showed an enhancement of the IFN-c PY-STAT3 response that was similar to that of CNTF (Fig. 2c).

Fig. 2 Concentration and time dependence of CNTF pre-treatment on IFN-c STAT3 response. (a) SH-SY5Y cells were pre-treated for 5 h with CNTF (0, 0.1, 1.0 or 100 pM) then challenged with (upper 3 panels) or without (lower panel) IFN-c 100 pM for 30 min. Blots were probed sequentially with anti-PY-STAT3 antibodies, anti-PY-STAT1 antibodies and anti-STAT3 antibodies. (b) SH-SY5Y cells were pre-treated with CNTF (1 nM) for shorter durations (30–60 min) or longer durations (3–5 h) of time then challenged with IFN-c for 30 min. The cell lysates were processed for immunodetection of PY-STAT3 followed by STAT3 protein for normalization. The induction was calculated as the ratio of the PY-STAT3 level after CNTF pre-treatment level alone plus the PY-STAT3 level after IFN-c 30 min alone (equal to 1.0) to the IFN-c PY-STAT3 response after CNTF pre-treatment (n ¼ 6–10 independent determinations). (c) SH-SY5Y cells were pre-treated for 5 h with LIF (1 nM) or CNTF (1 nM) or no pre-treatment (lanes 3 and 6) followed with no further treatment (left panels) or with IFN-c (100 pM) for 60 min (right panels). Cell lysates were collected and processed using ECL for detection of PY-STAT3, PY-STAT1 and STAT3.


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

CNTF induction of interferon-c STAT3 signaling 441

Induction of an IFN-c PY-STAT3 response in rat sympathetic neurons To determine whether the induction of PY-STAT3 by IFN-c also occurs in primary mammalian nerve cells, we utilized cultures of embryonic rat superior cervical ganglion neurons. IFN-c, through activation of STAT1, has previously been shown to block the dendrite outgrowth response of sympathetic neurons to BMP-7 (Kim et al. 2002). Ganglion neurons showed a selective PY-STAT3 increase in response to CNTF, which remained elevated even after 5-h incubation (Fig. 3) as compared with the desensitizing CNTF response of SH-SY5Y cells. IFN-c produced a selective PY-STAT1 response without a significant increase in PY-STAT3 (Fig. 3). Following pre-treatment of sympathetic neurons for 5 h with CNTF, acute exposure to IFN-c caused a 1.5-fold increase in the PY-STAT3 response over that of CNTF alone (Fig. 3b). The fold-increase in the IFN-c STAT3 response induced by CNTF was less than that seen in the SH-SY5Y cell line (compare with Fig. 1), however, this calculation was significantly influenced by the greater residual CNTF-mediated PY-STAT3 levels observed in the sympathetic neurons. Qualitatively the results from the primary neurons was similar to that from the cell lines, as prior to CNTF treatment sympathetic neurons responded to IFN-c with a restricted STAT1 signal but after CNTF treatment IFN-c was able to signal through a combination of STAT1 and STAT3. Therefore, the enhancement of PY-STAT3 by IFN-c after CNTF treatment was a common feature of primary nerve cells as well as neuroblastoma derived cell lines. Enhancement in IFN-c-mediated STAT nuclear activity after CNTF pre-treatment To determine whether IFN-c-induced STAT3 activation was of physiological relevance, we analyzed the induction of gene expression with the 3(G)3 luciferase reporter plasmid, 3(G)3 luc, transfected into HMN-1 murine hybrid motor neuronneuroblastoma cells. We used these cells because they are more efficiently transfected than are SH-SY5Y cells (Kaur and Halvorsen, data not shown). Pre-treatment of HMN-1 cells showed a CNTF induced increase in IFN-c mediated PY-STAT3 levels that was similar to that of SH-SY5Y cells (Fig. 4a). IFN-c alone activated the transcription to a small extent, but following CNTF pre-treatment, IFN-c produced a 7.5-fold greater increase in luciferase activity compared with IFN-c alone (Fig. 4b). The induced IFN-c response was threefold greater than that of CNTF alone and about twofold greater than the sum of the responses from CNTF and IFN-c when given alone (Fig. 4b). We analyzed the STAT–DNA complexes induced by IFNc with and without CNTF pre-treatment by EMSA. The binding of STAT derived from nuclear extracts to 32P-labeled oligonucleotides containing STAT-binding domains is maximal at about 30 min of CNTF treatment and declines with desensitization (Kaur et al. 2002). CNTF treatment of

(a)

(b)

Fig. 3 CNTF induction of IFN-c PY-STAT3 response in rat sympathetic neurons. (a) Cultured rat superior cervical ganglia neurons were not treated or pre-treated with rat CNTF (50 ng/mL) for 5 h, cells were then rinsed with fresh media. After 1 h, cells were either not further treated or treated with rat IFN-c (20 ng/mL) for 1 h. Cell lysates were processed as described for Fig. 1 and probed sequentially with antiPY-STAT3, anti-PY-STAT1 and anti-STAT3 antibodies. (b) Data from experiments as described in (a) were quantified and normalized against total STAT3 protein levels in lysates from each group. Data from three determinations were combined after normalization of each experiment to the PY-STAT3 levels observed after CNTF pre-treatment and IFN-c challenge (100%). No changes were observed in the IFN-c PY-STAT1 responses following CNTF pre-treatment.

HMN-1 cells, like SH-SY5Y cells, for 30 min forms at least three types of DNA-STAT complexes: those with STAT3 only, STAT3 and STAT1, and STAT1 only (data not shown and Kaur et al. 2002). After 7 h of continuous CNTF exposure complexes containing STAT1 were undetectable and STAT3 homomeric complexes were seen at reduced intensity (Fig. 4c). Acute treatment with IFN-c alone resulted in only a single type of STAT binding complex, which was


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

442 N. Kaur et al.

Fig. 4 Effect of CNTF pre-treatment on transcriptional activity. (a) Hybrid HMN-1 cells were untreated or pre-treated with CNTF (1 nM) for 7 h and challenged for 30 min with IFN-c. Cell lysates were subjected to immunoblot processing as described in Fig. 1. (b) HMN-1 cells were transfected with the luciferase reporter 3(G3) luc and the cotransfection control RSV-b gal plasmids. After 48 h, cells were treated with CNTF (1 nM) for 7 h and challenged with IFN-c (100 pM). Cells were harvested after 36 h and analyzed for luciferase and b-galactosidase activity. Data are presented as luciferase activity normalized to b-galactosidase activity (n ¼ 3 independent determinations). (c) Cultures of HMN-1 cells were treated as in (a) and nuclear extracts were subjected to EMSA. Some binding reactions included STAT1 or STAT3 antibodies as indicated (anti-STAT Ab). The bars labeled a, b, and c refer to STAT-DNA complexes with STAT3 only, STAT1/STAT3 together, and STAT1 only, respectively. The asterisk indicates a supershifted complex containing STAT3.

when cells were treated with CNTF and allowed to desensitize or treated with IFN-c alone for 30 min, represented a transcription complex not observed when cells were exposed to either agent alone under these conditions.

blocked by anti-STAT1 antibodies and not affected by antiSTAT3 antibodies (Fig. 4c). Treatment of HMN-1 cells with CNTF (1 nM) for 7 h followed by IFN-c (100 pM) treatment for 30 min showed a triplet of STAT1 and STAT3 containing complexes, one with STAT1 only, a second with STAT3 only, and a third with both STAT1–STAT3. The complex with STAT1–STAT3, absent

Dependence of PY-STAT3 induction on protein synthesis and protein kinases Stimulation of gp130 receptors in selected cell types is capable of activating various signaling pathways downstream of Jak. We tested the effect of individually blocking these different pathways on the CNTF induction of the IFN-c STAT3 response in SH-SY5Y cell with selective inhibitors. Treatment of SH-SY5Y cells prior to CNTF either with U0126, a MAPK kinase inhibitor, or BIM, a protein kinase C inhibitor, each blocked about 75% of CNTF-mediated enhancement of the IFN-c PY-STAT3 response (Fig. 5a). Neither H7, a less specific serine/ threonine kinase inhibitor, nor LY294002, an inhibitor of PI-3 kinase, inhibited the PY-STAT3 enhancement (Fig. 5a). Therefore, inhibition of either MAPK kinase or protein kinase C significantly decreased the CNTF-mediated IFN-c PY-STAT3 induction. To determine if new protein synthesis was required for CNTF-mediated induction we tested two different protein synthesis inhibitors. Incubation of SH-SY5Y cells with either emetine or puromycin prior to CNTF exposure prevented the enhancement of the IFN-c PY-STAT3 response (Fig. 5b). Thus ongoing protein synthesis was required for the CNTF induction response. Pre-treatment activation through the gp130 STAT binding domain, and STAT1 activity, are required for induction of IFN-c PY-STAT3 response. To further define the mechanism for gp130 cytokinemediated induction of the interferon STAT3 response, we used H-35 cell lines permanently expressing chimeric G-CSF/gp130 subunits with mutated STAT binding sites (133Y3F), mutated SHP binding sites (133Y2F) or with both


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

CNTF induction of interferon-c STAT3 signaling 443

(a)

(a)

(b)

(b) Fig. 5 Effects of protein kinase and protein synthesis inhibition on induction of IFN-c STAT3 response. (a) The ability of CNTF to induce an IFN-c PY-STAT3 response was determined in SH-SY5Y cells after treatment with the protein kinase inhibitors, BIM (10 lM), H7 (200 lM), LY294002 (16 lM) and U0126 (1 lM). SH-SY5Y cells were pre-treated with control media or media supplemented with the indicated protein kinase inhibitor for 1 h, then treated with CNTF (1 nM) for 5 h, IFN-c (100 pM) for 30 min or induced by CNTF (1 nM) for 5 h followed by IFN-c (100 pM) for 30 min. Cell lysates were processed for immunoblot analysis as described in Fig. 1(a) and PY-STAT3 signals were quantified and normalized against total STAT3 or STAT6 protein levels. The total CNTF-mediated IFN-c PY-STAT3 induction in control (not treated with protein kinase inhibitors) was set as 100% and the effect of drugs are presented as a percentage of that maximum PY-STAT3 induction. Data presented are the mean of n ¼ 3–5 independent experiments. (b) SH-SY5Y cells were treated with control media or media supplemented with following protein synthesis inhibitors for 1 h and cell lysates were processed as in (a). Immunoblots probed with anti-PY STAT3, anti-PY STAT1 and anti-STAT6 antibodies are indicated. Protein synthesis inhibitors used were 15 lg/mL puromycin and 10 lg/ mL emetine.

binding sites functional (133WT, Lai et al. 1995; Blanchard et al. 2001). Cells expressing the truncated, but functional gp130, showed the expected PY-STAT and MAPK responses after G-CSF stimulation and, further, G-CSF pre-treatment resulted in an enhanced IFN-c PY-STAT3 response (Fig. 6a, 133WT). However, cells with the mutated G-CSF/gp130 STAT binding domain showed no enhancement of the IFN-c STAT3 response (Fig. 6a, 133Y3F). Cells expressing mutant G-CSF/gp130 SHP-2 domains showed an 85% inhibition of the IFN-c enhanced PY-STAT3 response (Fig. 6a, 133Y2F). These results indicate that signaling of gp130 through both the STAT and the SHP-2 binding domains is required for induction of the STAT3 response. To determine if the lack of IFN-c STAT3 induction in the 133Y3F mutant cells could be assigned to either STAT1 or STAT3 specifically, HepG2 cells were transfected with

Fig. 6 Role of SHP, STAT binding domains and STAT1 activity for induction of IFN-c PY-STAT3 response. (a) H-35 hepatoma cells stably expressing the chimeric G-CSF/gp130 subunits composed of functional STAT and SHP-2 binding domains [(133)WT], with chimeras lacking SHP-2 binding site [(133)Y2F] or chimeras lacking STAT binding site [(133)Y3F] were treated with G-CSF (75 ng/mL) for 5 h and then challenged with rat IFN-c (2 ng/mL) for 30 min. The cell lysates were processed for immunodetection of PY-STAT3 followed by STAT3 protein for normalization. The total G-CSF-mediated PY-STAT3 induction in [(133)WT] was 100% and the percentage of that PY-STAT3 induction is shown (n ¼ 3 independent determinations). (b) HepG2, hepatoma cells were transiently transfected with the empty vector as control or dominant negative STAT1F mutant. Cells were untreated (control) or pre-treated with only IL-6 (0.5 nM) for 5 h or with only IFN-c (200 pM) for 30 min or both IL-6 (0.5 nM) for 5 h followed by IFN-c (200 pM) for 30 min Results of immunoblotting with anti-PY-STAT3 and anti-STAT3 antibodies are shown.

an expression vector for the dominant negative STAT1F (Phe substitution for Tyr701) or the empty vector as a control (Nakajima et al. 1996). Cells transfected with a mutant STAT1 showed a reduced IL-6-mediated induction of an IFN-c STAT3 response as compared with vector only transfected cells (Fig. 6b). Thus, overexpression of a dominant-negative STAT1 partially prevented the IFN-c STAT3 induction. Testing for the role of STAT3 using a dominant negative STAT3 construct was complicated by the need to test for an IFN-c-mediated PY-STAT3 response. These results suggest that STAT1, at least, is one of the signaling molecules that requires the STAT binding site which is mutated in G-CSF/gp130(133Y3F) receptor chimeras. There-


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

444 N. Kaur et al.

fore, gp130 cytokines require both the SHP-2 and the STATmediated signaling pathways for optimal induction of the IFN-c PY-STAT3 response. Discussion

Cells are exposed to various combinations of cytokines and growth factors during normal development and during periods of trauma and infection. The gp130-related cytokines function as nerve growth factors, neural differentiation factors and neural protection factors following axotomy (Adler et al. 1979; Nishi and Berg 1981; Sendtner et al. 1990; Mitsumoto et al. 1994; Koshlukova et al. 1996; Finn et al. 1998). IFN-c is induced as a proinflammatory cytokine in the nervous system during viral infection, trauma and in some neurodegenerative diseases (Traugott and Lebon 1988; Panitch 1992; Lau and Yu 2001). Here we have described a functional interaction between these two families of extracellular signaling molecules, both of which signal through Jak/STAT pathways. Prior exposure of cells to gp130 cytokines such as CNTF and IL-6 cause a significant increase in the IFN-c-mediated activation of STAT3, while the activation of STAT1 remains unchanged. This induction of an IFN-c STAT3 response was observed in the human neuroblastoma cell line, SH-SY5Y, a mouse hybrid motor neuron/neuroblastoma line, HMN-1 and two different hepatoma cell lines, HepG2 and H-35. The change in IFN-c response after pre-treatment with CNTF was very different from the effect on other gp130-related cytokines. We have previously shown that continuous exposure of most cell types to CNTF or LIF induces desensitization to further stimulation by all other members of the CNTF cytokine family (Kaur et al. 2002). Thus, while the STAT1 and STAT3 responses to CNTF cytokines were desensitized, the IFN-c-mediated STAT1 response was unchanged and the STAT3 response is enhanced. CNTF induced an IFN-cmediated PY-STAT3 effect in primary neurons of rat sympathetic ganglia also. However, the amount of the increase in PY-STAT3 signal was not as great in the primary nerve cells as was seen in the cell lines. In part this was due to the relatively small amount of desensitization seen in primary nerve cells during prolonged exposure to CNTF resulting in a very strong residual PY-STAT3 signal (Guo et al. 1999; Kaur et al. 2002). The smaller response in sympathetic neurons could also reflect differences in IFN-c responses between different cells in their ability to be induced by prolonged CNTF. Cellular responses to IFN-c are largely mediated by activation of STAT1, however, other pathways can be activated. Indeed only rarely IFN-c responses are associated with STAT3 activation (Darnell et al. 1994; Boehm et al. 1996; Stark et al. 1998). IFN-c cell responses have even been described in the complete absence of STAT1 (Ramana et al. 2001). Such exceptions are found in human neutrophils

that have a STAT1 and STAT3 response, while eosinophils and monocytes show only a STAT1 response (Caldenhoven et al. 1999). Adipocytes also display both a STAT1 and STAT3 response to IFN-c (Stephens et al. 1998). The mechanism for such cell type-specific activation of STAT3 is not known. The work presented here supports the possibility that interaction with other types of cytokines can induce such a change in cellular response. Results from a survey of pharmacological inhibitors of signaling pathways associated with CNTF revealed that block of both the MAPK pathway by U1026 and the protein kinase C pathway by BIM inhibited CNTF-mediated enhancement of the IFN-c PY-STAT3 response. Many cells respond to CNTF-related cytokines by activation of MAPK including the cell lines used in this study (Stahl et al. 1994; Inoue et al. 1996; Taga 1996; Lai et al. 1999). However, protein kinase C is not typically activated by these cytokines. A possible link between cytokine signaling and protein kinase C activity has been previously reported. Maximal activation of MAPK by the gp130 cytokine, LIF, in 3T3-L1 cells was shown to require protein kinase C activity (Schiemann and Nathanson 1994). Therefore the effects of both U1026 and BIM may have been via inactivation of a MAPK. A requirement for MAPK signaling in the enhancement of the IFN-c response was further supported by the studies with chimeric G-CSF/gp130 receptor expressing H-35 cells. These results showed that deletion of the box 3 STAT binding sites or the SHP-2 binding site inhibited the IFN-c STAT3 induction response. Gp130-mediated MAPK signaling requires a functional SHP-2 binding site as does the tyrosine phosphatase, SHP-2 (for review see Taga 1996). It is interesting to note that while cells expressing the mutated SHP-2 site showed inhibited IFN-c STAT3 signaling, the G-CSF-mediated STAT3 response was enhanced. This enhanced G-CSF STAT3 response was probably due to the reduced rate of PY-STAT3 dephosphorylation in the absence of recruited SHP-2. Thus STAT3 signaling overall was not negatively impacted by the mutated gp130, only that mediated by IFN-c. We do not yet know the target of the MAPK activity. Previous studies have shown that STAT3 undergoes serine phosphorylation at S727 upon cytokine stimulation and that this phosphorylation affects its function (Wen et al. 1995; Zhang et al. 1995). However in SH-SY5Y cells this serine phosphorylation is inhibited by H7 but not by MAPK kinase or protein kinase C inhibitors (Boulton et al. 1995; Malek and Halvorsen 1997). Our observation that IFN-c STAT3 induction was inhibited by U1026 and BIM does not support a mechanism whereby a STAT3 modification at S727 affects its ability to be activated by IFN-c. The IFN-c receptor could be a substrate for MAPK, perhaps at the STAT1 binding site, resulting in an increased affinity for STAT3 binding. However, a mechanism involving covalent modification of


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

CNTF induction of interferon-c STAT3 signaling 445

STAT binding to the IFN-c receptor has not previously been described. Two lines of evidence support the need for new protein synthesis in the induction of the IFN-c STAT3 response. The blockade by puromycin and emetine suggests either a shortlived protein or a new protein is required for STAT3 induction. Further, the lack of induction in STAT-defective gp130 chimera receptor expressing cells suggests that a STAT transcriptional activity is required. The putative protein synthesized was apparently not the IFN-c receptor or STAT3 itself. There was no increase in the IFN-c STAT1 response, suggesting no increase in receptor numbers. Further, we saw no significant increase in total cellular STAT3 levels following pre-treatment with gp130 cytokines. The putative induced protein may well have a corequirement for both MAPK and STAT activity. The inhibition of induction in cells expressing the dominant negative STAT1 construct indicated that STAT1 was important but we could not determine if STAT3 was also needed as use of a dominant-negative STAT3 would preclude observing a STAT3 response by IFN-c. Thus, it may be that the cross-talk between gp130 cytokines and IFN-c involves activation of a gene that requires both STAT and MAPK, and this gene product is necessary to promote STAT3 interaction with the IFN-c receptor. STAT3 has a unique role among the STATs in regulating homeostasis (reviewed in Schindler 2002). While animals with functional mutations in the other STATs exhibit restricted phenotypes, STAT3 negative mice do not survive embryogenesis (Takeda et al. 1997). Induction of the STAT3 response in IFN-c signaling could be a mechanism to promote a STAT3 response in selected cells as a function of the composition of growth factors and cytokines. The overall impact of the IFN-c STAT3 response in induced cells is difficult to assess in this system. In the absence of CNTFcytokine pre-treatment, we were unable to detect IFN-cmediated STAT3 containing DNA binding complexes, even in cases where a small PY-STAT3 signal was observed, and there was only a minimal IFN-c effect on stimulating the STAT-driven reporter gene. However, there was an IFN-cmediated increase in STAT-driven gene expression and the appearance of a new STAT1/3 heterodimer after CNTFcytokine pre-treatment. These results suggest that in the absence of pre-treatment with gp130 cytokines, the IFN-c STAT3 response was much less efficient. In sympathetic neurons, which show a selective STAT3 response to CNTF, the stimulation of a STAT 1/3 heterodimer by IFN-c would represent a unique complex, one not produced without the prior exposure to CNTF. We have described here an example of cross-talk between the IFN-c and gp130 cytokines. Two families of cytokines that are likely to be expressed together in tissues under highly controlled situations during development or as a result of trauma and disease. The resulting interaction could result in important modifications to the IFN-c response associated

with genes uniquely regulated by STAT3 or combinations of STAT1 and STAT3. The work also suggests sites of regulation of IFN-c signaling that control STAT response which are as yet undefined. Acknowledgements We thank Regeneron Pharmaceuticals for recombinant human CNTF and Dr Heinz Baumann of Roswell Park Cancer Institute for the transfected H-35 cell lines. We are grateful to Ann Wohlhueter for technical support in the initial phases of the project. This work was supported by grants from the National Institutes of Health (NS30232 to SWH) and the National Science Foundation (IBN9728881 to SWH) and (IBN0121210 to DH).

References Adler R., Landa K., Manthorpe M. and Varon S. (1979) Cholinergic neurotrophic factors: intraocular distribution of soluble trophic activity for ciliary neurons. Science 204, 1434–1436. Blanchard F., Wang Y. P., Kinzie E., Duplomb L., Godard A. and Baumann H. (2001) Oncostatin M regulates the synthesis and turnover of gp130, leukemia inhibitory factor receptor alpha, and oncostatin M receptor beta by distinct mechanisms. J. Biol. Chem. 276, 47038–47045. Boehm S., Huck S. and Freissmuth M. (1996) Involvement of a phorbol ester-insensitive protein kinase C in the a2-adrenergic inhibition of voltage-gated calcium current in chick sympathetic neurons. J. Neurosci. 16, 4596–4603. Boehm U., Klamp T., Groot M. and Howard J. C. (1997) Cellular responses to interferon-gamma. Ann. Rev. Immunol. 15, 749–795. Boulton T. G., Zhong Z., Wen Z. L., Darnell J. E., Stahl N. and Yancopoulos G. D. (1995) STAT3 activation by cytokines utilizing gp130 and related transducers involves a secondary modification requiring an H7-sensitive kinase. Proc. Natl Acad. Sci. USA 92, 6915–6919. Caldenhoven E., Buitenhuis M., van Dijk T., Raaijmakers J., Lammers J., Koenderman L. and de Groot R. (1999) Lineage-specific activation of STAT3 by interferon-gamma in human neutrophils. J. Leukoc. Biol. 65, 391–396. Darnell J. E., Kerr I. and Stark G. (1994) Jak-STAT pathways and transcriptional activation in response to interferons and other extracellular signaling proteins. Science 264, 1415–1420. Ernsberger U., Sendtner M. and Rohrer H. (1989) Proliferation and differentiation of embryonic chick sympathetic neurons: effects of CNTF. Neuron 2, 1275–1284. Finn T. P., Kim S. and Nishi R. (1998) Overexpression of ciliary neurotrophic factor in vivo rescues chick ciliary ganglion neurons from cell death. J. Neurobiol. 34, 283–293. Guo X., Chandrasekaran V., Lein P., Kaplan P. L. and Higgins D. (1999) Leukemia inhibitory factor and ciliary neurotrophic factor cause dendritic retraction in cultured rat sympathetic neurons. J. Neurosci. 19, 2113–2121. Halvorsen S. W. and Berg D. K. (1989) Specific down-regulation of the a-bungarotoxin binding site on chick autonomic neurons by ciliary neurotrophic factor. J. Neurosci. 9, 3673–3680. Higgins D., Lein P. J., Osterhout D. J. and Johnson M. I. (1991) Tissue Culture of Mammalian Autonomic Neurons, in Culturing Nerve Cells (Banker, G., Goslin, K., eds), pp. 177–205. MIT Press, Cambridge, MA. Hirota H., Yoshida K., Kishimoto T. and Taga T. (1995) Continuous activation of gp130, a signal-transducing receptor component for


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

446 N. Kaur et al.

interleukin 6-related cytokines, causes myocardial hypertrophy in mice. Proc. Natl Acad. Sci. USA 92, 4862–4866. Hu X. Y., Herrero C., Li W. P., Antoniv T. T., Falck-Pedersen E., Koch A. E., Woods J. M., Haines G. K. and Ivashkiv L. B. (2002) Sensitization of IFN-gamma Jak-STAT signaling during macrophage activation. Nature Immunol. 3, 859–866. Ihle J. N. (1996) STATs: signal transducers and activators of transcription. Cell 84, 331–334. Inoue M., Nakayama C. and Noguchi H. (1996) Activating mechanism of CNTF and related cytokines. Molec. Neurobiol. 12, 195–209. Kaur N., Wohlhueter A. L. and Halvorsen S. W. (2002) Activation and inactivation of signal transducers and activators of transcription by ciliary neurotrophic factor in neuroblastoma cells. Cell. Signalling 14, 419–429. Kim I.-J., Beck H. N., Lein P. J. and Higgins D. (2002) Interferon gamma induces retrograde dendritic retraction and inhibits synapse formation. J. Neurosci. 22, 4530–4539. Kordula T., Bugno M., Goldstein J. and Travis J. (1995) Activation of signal transducer and activator of transcription-3 (STAT3) expression by interferon-c and interleukin-6 in hepatoma cells. Biochem. Biophys. Res. Commun. 216, 999–1005. Koshlukova S., Finn T. P., Nishi R. and Halvorsen S. W. (1996) Identification of functional receptors for ciliary neurotrophic factor on chick ciliary ganglion neurons. Neuroscience 72, 821–832. Kotzbauer P., Lampe P., Estus S., Milbrandt J. and Johnson E. (1994) Postnatal development of survival responsiveness in rat sympathetic neurons to leukemia inhibitory factor and ciliary neurotrophic factor. Neuron 12, 763–773. Kurek J. B., Austin L., Cheema S. S., Bartlett P. F. and Murphy M. (1996) Up-regulation of leukaemia inhibitory factor and interleukin-6 in transected sciatic nerve and muscle following denervation. Neuromusc. Disord. 6, 105–114. Lai C.-F., Ripperger J., Morella K. R., Wang Y. P., Gearing D. P., Fey G. H. and Baumann H. (1995) Separate signaling mechanisms are involved in the control of STAT protein activation and gene regulation via the interleukin 6 response element by the Box 3 motif of gp130. J. Biol. Chem. 270, 14847–14850. Lai C.-F., Ripperger J., Wang Y., Kim H., Hawley R. B. and Baumann H. (1999) The STAT3-independent signaling pathway by glycoprotein 130 in hepatic cells. J. Biol. Chem. 274, 7793–7802. Lamb P., Seidel H. M., Haslam J., Milocco L., Kessler L. V., Stein R. B. and Rosen J. (1995) Stat protein complexes activated by interferongamma and gp130 signaling molecules differ in their sequence preferences and transcriptional induction properties. Nucleic Acids Res. 23, 3283–3289. Lau L. T. and Yu A. C. (2001) Astrocytes produce and release interleukin-1, interleukin-6, tumor necrosis factor alpha and interferongamma following traumatic and metabolic injury. J. Neurotrauma 18, 351–359. Leonard W. J. and Oshea J. J. (1998) Jaks and STATs – biological implications. Ann. Rev. Immunol. 16, 293–322. Levison S. W., Ducceschi M. H., Young G. M. and Wood T. L. (1996) Acute exposure to CNTF in vivo induces multiple components of reactive gliosis. Exp. Neurol. 141, 256–268. Mahboubi K. and Pober J. S. (2002) Activation of signal transducer and activator of transcription 1 (STAT1) is not sufficient for the induction of STAT1-dependent genes in endothelial cells – comparison of interferon-gamma and oncostatin M. J. Biol. Chem. 277, 8012–8021. Malek R. L. and Halvorsen S. W. (1997) Opposing regulation of ciliary neurotrophic factor receptors on neuroblastoma cells by distinct differentiating agents. J. Neurobiol. 32, 81–94. Metcalf D. (2003) The unsolved enigmas of leukemia inhibitory factor. Stem Cells 21, 5–14.

Mitsumoto H., Ikeda K., Klinkosz B., Cedarbaum J., Wong V. and Lindsay R. (1994) Arrest of motor neuron disease in wobbler mice cotreated with CNTF and BDNF. Science 265, 1107–1110. Nakajima K., Yamanaka Y., Nakae K., Kojima H., Ichiba M., Kiuchi N., Kitaoka T., Fukada T., Hibi M. and Hirano T. (1996) A central role for Stat3 In IL-6-induced regulation of growth and differentiation in M1 leukemia cells. EMBO J. 15, 3651–3658. Nishi R. and Berg D. K. (1981) Two components from eye tissue that differentially stimulate growth and development of ciliary ganglion neurons in culture. J. Neurosci. 1, 505–513. Orlovsky K., Theodor L., Malovani H., Chowers Y. and Nir U. (2002) Gamma interferon down-regulates Fer and induces its association with inactive Stat3 in colon carcinoma cells. Oncogene 21, 4997–5001. Panitch H. S. (1992) Interferons in multiple sclerosis: a review of the evidence. Drugs 44, 946–962. Pennica D., King K., Shaw K., Luis E., Rullamas J., Luoh S.-M., Darbonne W., Knutzon D., Yen R., Chien K., Baker J. and Wood W. I. (1995) Expression cloning of cardiotrophin 1, a cytokine that induces cardiac myocyte hypertrophy. Proc. Natl Acad. Sci. USA 92, 1142–1146. Ramana C. V., Gil M. P., Han Y. L., Ransohoff R. M., Schreiber R. D. and Stark G. R. (2001) Stat1-independent regulation of gene expression in response to IFN-gamma. Proc. Natl Acad. Sci. USA 98, 6674–6679. Salazar-Grueso E. F., Kim S. and Kim H. (1991) Embryonic mouse spinal cord motor neuron hybrid cells. Neuroreport 2, 505–508. Schiemann W. P. and Nathanson N. M. (1994) Involvement of protein kinase C during activation of the mitogen-activated protein kinase cascade by leukemia inhibitory factor: evidence for participation of multiple signaling pathways. J. Biol. Chem. 269, 6376–6382. Schindler C. (2002) JAK-STAT signaling in human disease. J. Clin. Invest. 109, 1133–1137. Sendtner M., Kreutzberg G. and Thoenen H. (1990) Ciliary neurotrophic factor prevents the degeneration of motor neurons after axotomy. Nature 345, 440–441. Shuai K., Stark G. R., Kerr I. M. and Darnell J. E. Jr (1993) A single phosphotyrosine residue of Stat91 required for gene activation by interferon-gamma. Science 261, 1744–1746. Stahl N., Boulton T., Farruggella T., Ip N., Davis S., Witthuhn B., Quelle F., Silvennoinen O., Barberi G., Pellegrini S., Ihle J. and Yancopoulos G. (1994) Association and activation of Jak-Tyk kinases by CNTF-LIF-OSM-IL6 b receptor components. Science 263, 92–95. Stahl N., Farruggella T., Boulton T., Zhong Z., Darnell J. Jr and Yancopoulos G. (1995) Choice of STATs and other substrates specified by modular tyrosine-based motifs in cytokine receptors. Science 267, 1349–1353. Stark G. R., Kerr I. M., Williams B. R. G., Silverman R. H. and Schreiber R. D. (1998) How cells respond to interferons. Annu. Rev. Biochem. 67, 227–264. Stephens J. M., Lumpkin S. J. and Fishman J. B. (1998) Activation of signal transducers and activators of transcription 1 and 3 by leukemia inhibitory factor, oncostatin-m, and interferon-gamma in adipocytes. J. Biol. Chem. 273, 31408–31416. Sun Y., Landis S. C. and Zigmond R. E. (1996) Signals triggering the induction of leukemia inhibitory factor in sympathetic superior cervical ganglia and their nerve trunks after axonal injury. Mol. Cell. Neurosci. 7, 152–163. Symes A., Lewis S., Corpus L., Rajan P., Hyman S. E. and Fink J. S. (1994) STAT proteins participate in the regulation of the vasoactive intestinal peptide gene by the ciliary neurotrophic factor family of cytokines. Mol. Endocrin. 8, 1750–1763. Symes A., Corpus L. and Fink J. S. (1995) Differences in nuclear signaling by leukemia inhibitory factor and interferon-gamma – the


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447

CNTF induction of interferon-c STAT3 signaling 447

role of Stat proteins in regulating vasoactive intestinal peptide gene expression. J. Neurochem. 65, 1926–1933. Taga T. (1996) GP130, a shared signal transducing receptor component for hematopoietic and neuropoietic cytokines. J. Neurochem. 67, 1–10. Takeda K., Noguchi K., Shi W., Tanaka T., Matsumoto M., Yoshida N., Kishimoto T. and Akira S. (1997) Targeted disruption of the mouse STAT3 gene leads to early embryonic lethality. Proc. Natl Acad. Sci. USA 94, 3801–3804. Traugott U. and Lebon P. (1988) Interferon-gamma and Ia antigen are present on astrocytes in active chronic multiple sclerosis lesions. J. Neurol. Sci. 84, 257–264. Wells J. A. and deVos A. M. (1996) Hematopoietic receptor complexes. Annu. Rev. Biochem. 65, 609–634. Wen Z., Zhong Z. and Darnell J. E. (1995) Maximal activation of transcription by STAT1 and STAT3 requires both tyrosine and serine phosphorylation. Cell 82, 241–250. Winter C. G., Saotome Y., Levison S. W. and Hirsh D. (1995) A role for ciliary neurotrophic factor as an inducer of reactive gliosis, the glial

response to central nervous system injury. Proc. Natl Acad. Sci. USA 92, 5865–5869. Wishingrad M. A., Koshlukova S. and Halvorsen S. W. (1997) Ciliary neurotrophic factor stimulates the phosphorylation of two forms of STAT3 in chick ciliary ganglion neurons. J. Biol. Chem. 272, 19752–19757. Yuan J., Wegenka U. M., Lutticken C., Buschmann J., Decker T., Schindler C. W., Heinrich P. C. and Horn F. (1994) The signalling pathways of interleukin-6 and gamma interferon converge by the activation of different transcription factors which bind to common responsive DNA elements. Mol. Cell. Biol. 14, 1657–1668. Zhang X. F., Blenis J., Li H.-C., Schindler C. and Chen-Kiang S. (1995) Requirement of serine phosphorylation for formation of STATpromoter complexes. Science 267, 1990–1994. Zhou H., Summers S. A., Birnbaum M. J. and Pittman R. N. (1998) Inhibition of Akt kinase by cell-permeable ceramide and its implications for ceramide-induced apoptosis. J. Biol. Chem. 273, 16568–16575.


2003 International Society for Neurochemistry, J. Neurochem. (2003) 87, 437–447